Parsing of non unique tags using xmllint in bash

I have an xml in this format

      <RECORD> 
          <RUN_TEST_SUCCESS> 
               <TEST_NAME>test1 </TEST_NAME> 
          </RUN_TEST_SUCCESS> 
      </RECORD> 
       <RECORD> 
            <RUN_TEST_FAILURE> 
                <TEST_NAME> test2</TEST_NAME> 
                <FILE_NAME> eg.c </FILE_NAME> 
                <LINE_NUMBER> 478 </LINE_NUMBER> 
                <CONDITION> condition1</CONDITION> 
             </RUN_TEST_FAILURE> 
       </RECORD>
...... 

How do i parse only run_test_failure tag and get the information inside it using xlint?

I have tried xlint --xpath 'string(/RECORD[2]/RUN_TEST_FAILURE/TEST_NAME)' myxml.xml gives me the correct value , but i do not know the position 'i' (here 2) for the failures. I only have the number of run_test_failure tags information. How to iterate of all the record tags, check if "run_test_failure" tag is present and then extract information?